Spotify songs dataset

General info

This plot shows how many tracks are in a particaular music genre.
The genre with the largest number of tracks is electro.

The graph indicates how many different genres and sungnres there are in dataset.
It can be noted that each genre has 4 subgenres.


This plot shows how many musicians perform in a particaular music genre.
It can clearly be seen that genres are equally popular among the musicians.
The genre with the largest number of performers is electro

The graph indicates how many different artists released their tracks in each year.
It can be noted that spotify is more focused on songs released over the past 3 years.


This plot shows the distribution of the most interesting and sophostocated parameters in dataset.
danceability - how suitable a track is for dancing
spechiness - the presence of spoken words in a track
energy - a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity.
Danceability outliers - probably rap or rock, spechiness outliers - probably rap.


Hypotheses for common values

1. Different genres have different speed rate. The highest speed has electro.

The hypothesis is true. It can clearly be seen that median for edm is the highest(=127). Rock took the second place


2. Most popular are tracks released
in the era of the greatest popularity of the genre.

3. An artist is most popular in one genre.

This is quite true. It is can clearly be seen in rock distribution.
While rock was the superior genre (70-80s), it had high level of popularity.
This graph also shows the history of music development.

This hypothesis is not correct. To achieve success, it is not necessary to develop in one genre.
There are many successful artists, with a high level of popularity for different genres.


Hypotheses for more interesting values

1. Rap has the highest speechiness.

2. Danceability and speechiness are interdependent

Rap does have the highest median in speechiness.
Also it has many outliers with great value.

This hypothesis is not correct. These parameters are not related.


3. Danceability and energy are interdependent.

4. Danceability and popularity are interdependent

These parameters are not related to each other.


5. Danceability and instrumenatlness are interdependent.

6. Nowadays the music is more danceable than before.

These parameters are not related.

Music became less danceable.


Interesting (funny) hypotheses

1. The popularity is higher for earlier tracks.

2. Remixes have lower popularity.

This is possibly true.

It is definitely true for some genres(for example, rock)